Figure 1—figure supplement 2.

(A) Comparison of swimming speed, vertical position in the water column (Z position), and social distance in groups of five fish at room temperature (RT; 25 – 26 °C) and at 36 °C. Fish were kept at the indicated temperature for 15 min. Each datapoint represents the mean value during a one-minute time window (10 time windows in total). Box plots show median and 25th and 75th percentiles, circles and error bars indicate mean and s.d. over time windows. Swimming trajectories were analyzed in two dimensions using DeepLabCut. Z position is normalized to the distance between the bottom of the tank (0) and the water surface (1). Social distance is the mean distance between all pairs of fish in the group. (Swimming speed: p<0.001, Z position: p<0.001, p=0.53, Wilcoxon rank sum test). (B) Swimming speed of adult fish kept individually in 1 L tanks measured at 27 °C. Fish were either kept at 27 °C or at 36 °C for 7 days prior to measurements and received different treatments (no injection, buffer injection, or HSV1 injection). Plot symbols represent data from individual fish; box plots show median and 25th and 75th percentiles, circles and error bars indicate mean and s.d. over individual fish. N: number of fish. No significant differences were observed (p>0.54; ANOVA test). (C) Sequence of virus injections, temperature changes and associative conditioning to assess effects of experimental manipulations on discrimination learning. (D) Learning index (behavioral discrimination score) on the last day of training. Plot symbols represent data from individual fish; box plots show median and 25th and 75th percentiles, circles and error bars indicate mean and s.d. over individual fish. N: number of fish. Control fish (group 1) were not injected and kept at standard laboratory temperature. The experimental group (group 2) was injected with HSV1[LTCMV:jGCaMP7b], an amplicon type of HSV1 with an insert encoding the calcium indicator GCamp7b (Dana et al., 2019) under the control of a non-specific promoter for long-term expression (LTCMV), and subsequently kept at 36 °C for 2 days before training (A). Performance was not significantly different between groups (p=0.76, Wilcoxon rank sum test).